Breadcrumb
Search and Filter
The Repository
Other
Research synthesis is AI-generated, human reviewed. Updated 11/2026.
Displaying 1 - 30 of 292
Early Prediction Of AI-Assisted Cheating Risk In Online Exams Through Learning Analytics
Gokhan Akcapınar. (09/2026). arXiv.
Logistic regression predicted AI-assisted cheating risk from first-eight-week learning management system traces with 73.1% accuracy, a balanced accuracy of 0.718, and an area under the curve of 0.763.
Three Pathways Of Student-AI Interaction: Constraint-First Design For Higher-Order Thinking
Fatima Tuz Zahra. (09/2026). arXiv.
Undergraduate student interactions fell into three paths: 46% Passive Review, 18% Direct Question, and 36% Strategic Dialogue in primary classifications.
AI-Based Scoring Systematically Underestimates Conceptual Understanding Of Linguistically Weak Students' Explanations In Physics
Markus Sebastian Feser, Paul Tschisgale. (09/2026). arXiv.
Across all 11 AI scoring approaches, lower writing quality was associated with 1.3- to 3.9-times higher odds of underestimating students' conceptual understanding in physics.
Argus: Academic Integrity In The Era Of Generative AI
David Racovan, Ajay Rawat, Christopher K. May, Jeffrey A. Turkstra. (09/2026). arXiv.
In Spring 2026, 45.7% of enrolled students were flagged for suspected LLM-assisted code development following automated triage and manual review.
A Benchmark For LLM'S Understanding Of Middle School And High School Science Topics
Noah L. Schroeder, Yessy Eka Ambarwati, Yuji Zhang, ChengXiang Zhai. (09/2026). arXiv.
Most tested models scored above 90% accuracy on middle and high school standards, with larger model size not consistently predicting higher performance.
"We'Ll Fix It Later": Education, AI, And The Deferral Of Privacy In EdTech
Meghna Manoj Nair, Rachel Greenstadt. (09/2026). arXiv.
One-third of audited platforms made no meaningful AI disclosure despite visible AI features, and 73% provided only generic accountability and breach-response language.
Towards Synergistic Teacher-AI Interactions With Generative Artificial Intelligence
Mutlu Cukurova, Wannapon Suraworachet, Qi Zhou, Sahan Bulathwela. (09/2026). arXiv.
Among reviewed generative AI studies supporting teachers, operational teaming was most common (62.5%), followed by transactional (17.5%), situational (17.5%), and praxical teaming (2.5%).
A Workshop Series For Effective Use Of AI In Uncertain Times: Building A Physics Faculty Learning Community
David Perl-Nussbaum, Noah D. Finkelstein. (09/2026). arXiv.
Faculty most valued collectively airing a shared challenge, reviewing cross-course student data, and surfacing concerns, with all participants expressing interest in continuing the learning community.
AI-Written Admissions Essays Are Widespread But Penalized
Calvin Isley, Johann Gaebler, Sharad Goel. (09/2026). arXiv.
By 2025, 56% of applicants submitted at least one primarily AI-generated essay—reaching 69.3% among international applicants—despite an explicit prohibition against AI use.
Artificial Intelligence Driven Physics Assignments Using Context Prompts
Miguel Rodriguez, Peter Wulff. (09/2026). arXiv.
Context-prompted assignments helped students move beyond answer-seeking to connect introductory physics concepts directly to their jobs, hobbies, and everyday lives.
The Role Of Implicit And Explicit Demographic Signals In Large Language Model-Based Student Assessment
Donya Rooein, Luca Benedetto, Dirk Hovy. (09/2026). arXiv.
In essay scoring, implicit conversation history caused Llama-70B scores to inflate by 1.57 points relative to default, while explicit demographics caused small shifts.
Limits Of LLM Text Detectors In Education
Lukas Gehring, Benjamin PaaBen. (09/2026). arXiv.
Detector performance drops sharply on texts with intermediate assistance, particularly AI-assisted rewrites of human-written essays.
Generative AI Performance In Core Undergraduate Mathematics: A Curriculum-Level Case Study
Benjamin J. Walker, Nikoleta Kalaydzhieva, Beatriz Navarro Lameda, Ruth A. Reynolds. (09/2026). arXiv.
Generative AI achieved an overall average score of 71.1 out of 100, performing on par with the student cohort average of 69.1.
Eduguard: A Safe RAG-Based LLM Tutor For Programming Education
S M Asif Hossain, Ruksat Khan Shayoni, M. F. Mridha, Jungpil Shin. (09/2026). arXiv.
On the BILearn-CS benchmark, EduGuard achieved 90.1% correctness, 89.4% grounding, and reduced hallucination to 4.9% and direct-answer leakage to 9.8%.
The 5P Reflection Model For Education In The Generative Artificial Intelligence (Genai) Era
Rajan Kadel, Samar Shailendra, Islam Mohammad Tahidul, Urvashi Rahul Saxena, Aakanksha Sharma, Sabitra Kaphle. (09/2026). arXiv.
Traditional reflection models fail to account for generative AI's probabilistic nature, risks of superficial synthesis, and the need to verify learner authenticity.
When Saying No Makes Better Videos: Designing Dual Gatekeeping For Pedagogically Grounded AI Content Creation
Yearim Kim, Injun Baek, Nojun Kwak. (08/2026). arXiv.
Educator review based on multimedia learning principles increased ratings across all 12 principles, raising mean ratings from 3.07 to 3.86 on a 5-point scale.
When LLMs Slow Down: How Environmental Impacts Mediate University Students'LLM Usage
Hyeonwook Kim, Xuesi Chen, Alex Cabral, Cindy Kaiying Lin, Udit Gupta, Josiah Hester. (08/2026). arXiv.
Selection of lower-carbon modes dropped from about 45% in the fastest eco mode to under 5% in the slowest eco mode as response delays grew.
The AI Adaptation Gap In Higher Education: Students, Faculty, And Administrative Staff
Yuriy S. Braun, Salavat M. Khafizov. (08/2026). arXiv.
Students reported higher AI-use intensity and perceived usefulness than faculty and staff, while faculty and staff reported stronger academic integrity concerns and responsible-use norms.
When LLMs Slow Down: How Environmental Impacts Mediate University Students' LLM Usage
Hyeonwook Kim, Xuesi Chen, Alex Cabral, Cindy Kaiying Lin, Udit Gupta, Josiah Hester. (08/2026). arXiv.
Selection of eco modes dropped sharply from about 45% in mode 1 to below 5% in mode 5 as response delays increased.
AI, Expert Or Peer? Provider Biases And Feedback Uptake Among Pre-Service Teachers
Lucas Jasper Jacobsen, Ute Mertens, Thorben Jansen, Kira Elena Weber. (08/2026). arXiv.
Pre-service teachers rated large language model feedback more positively on fairness, usefulness, and acceptance when they mistakenly believed it came from a human.
Evaluation In The Age Of AI: Output As Evidence Of Learning
Md Zarzees Uddin Shah Chowdhury, Samin Rahman Khan. (08/2026). arXiv.
While 15 out of 20 respondents reported their institution relies heavily on AI detection software, only 4 had high confidence in detector accuracy and fairness.
ChatGPT Solves All Tested Qiskit Homework Assignments
Alexei Kaltchenko, Gurnivaj Tiwana. (08/2026). arXiv.
ChatGPT produced final artifacts that executed and passed autograding in all 150 tested sessions across the three introductory quantum programming assignments.
Large-Scale AI Grading Of Handwritten Physics Assessments: Score Agreement And Olympiad Team Selection Outcomes
Praveen Pathak, Siddharth Tiwary, Charudatt Kadolkar, Vijay Singh, David J. Rakestraw, Shirish Pathare, Anwesh Mazumdar. (08/2026). arXiv.
AI total scores tracked official examiner scores closely, achieving correlations between 0.91 and 0.97 across exams, and identified the exact same five-student Olympiad team as human examiners.
Learnai: Just-In-Time AI Co-Creation Across Disciplines At A University
Weihao Qu, Ling Zheng, Chris Buzaid, Daniel Crawford. (08/2026). arXiv.
Clients across disciplines co-created 36 portfolio websites and over 20 deployed web applications with peer tutors.
Revisiting The Performance Of Generative Artificial Intelligence On Introductory Object-Oriented Programming Assessments: Insights From 2026
Marina Lepp, Joosep Kaimre. (08/2026). arXiv.
All five AI systems achieved higher average scores than the student cohort on most assessments, with ChatGPT-5.2 scoring 31.01 and Claude reaching 30.95 out of 33 on the exam.
Mitigating AI Risks In Computing Education Via LLM-Driven Lecture Video Curation
Owen Tang, Alexandra Vassar, Jake Renzella. (08/2026). arXiv.
Proprietary models achieved near-parity with human lecturer ratings for relevance and sufficiency, despite having low exact timestamp overlap with the lecturer's selections.
Do Assessment Instruments Measure The Same Thing For Humans And LLMs? A Latent Structure Analysis
Alona Strugatski, Licol Zeinfeld, Giora Alexandron. (08/2026). arXiv.
Tests showed systematic differences between human and language model factor structures, indicating the exams may not measure the same constructs in humans and AI.
INSIDE the Student's Mind: Jointly Modeling Latent Reasoning and Action in LLM Student Simulators
Rose Niousha, Minwoo Kang, Narges Norouzi. (08/2026). arXiv.
Fine-tuning models on reconstructed internal dialogue achieved the closest match to real student code distributions and the highest reasoning alignment, reaching up to 57.9%.
Making AI-Generated Feedback Matter: A Large-Scale Study Of Feedback Workflows And Student Enactment
Omar Alsaiari, Nilufar Baghaei, Jason M. Lodge, Naomi Winstone, Dragan Gašević, Hassan Khosravi. (08/2026). arXiv.
Enacted Feedback was associated with significantly higher feedback uptake (estimated probability of 26.2%) compared with Directed Feedback (14.1%) and Self-Directed Feedback (0.1%).
Semantic Variability Of LLM-Generated Replies Across LLMs: Implications For Designing Conversation-Based Assessment
Jiangang Hao. (08/2026). arXiv.
Models generated replies more semantically similar to their own outputs (similarity 0.715 to 0.795) than to other models (0.443 to 0.604).
